I'm all for abandoning projects that makes no sense to maintain. But if this happens too often, then somewhere along the decision chain people are missing the mark on what's worth building and supporting. In Google's case, they should probably put a bit more thought into it before releasing something to the public. This situation, to me, is a sign of overconfidence among Google workers about their abilities. In humbler places, they do side projects too but before they start them there's a lot of discussion about the purpose of the project beyond the code and the value it brings. When I saw the list of projects Google had killed, most of those projects looked plain boring to work on and seemed like undergraduate projects. Then again, may be it's it's own strategy to build as many stuff as you can quickly hoping some of them might take off.